The major effort of this section continues to be the refining of existing methodologies for the study of cortical functioning based on positron emission tomography (PET) and to a lesser degree electrical brain mapping procedures particularly as they complement PET information. We continue to contribute to the development and comparison of three methods of data abstraction, the "peel", the "ring" and the region of interest (ROI). We have been examining the benefits of standardizing the abstracted PET data to emphasize that information that directly pertains to the activity of localized regions of the brain as opposed to the overall glucose metabolic rate (GMR) of the brain. To accomplish this we have explored multivariate analytical approaches other than that which has been previously used (repeated measures analysis of variance) which we feel is not suitable to the important biomedical questions that PET should be addressing. We have pursued as thorough an analysis as possible of the FDG PET representations of two behavioral paradigms, repeated somatosensory stimuli and an auditory continuous performance test. They complement each other in that in the former the "normal" response is to habituate to a "benign" sensory stimulus, whereas the latter task requires selective attention. Differences are observable and suggest that PET is a modality where specific functionally distinct regions of the frontal cortex can be examined. These paradigms are currently in use to study normal habituation and selective attention and to evaluate whether these fundamental processes are altered in mental illness. To date, schizophrenic patients do not show the diminution in electrical response to repetitive somatosensory stimuli observed in normals. These differences appear to be reflected in a different PET pattern of GMR.